Date post: | 28-Mar-2018 |
Category: |
Documents |
Upload: | vuongthuan |
View: | 216 times |
Download: | 4 times |
111
Rita Rafajlovičová
Presenting speech in literary and non-literary texts
Abstract
The paper describes an investigation into the use of different ways of presenting speech in a small size
corpus of contemporary written literary and non-literary texts. It provides a brief description of the
corpus, its size and sources the texts were selected from. In order to determine the distribution and
frequency of the different categories of speech presentation, quantitative analyses of the various
different text types were carried out. The theoretical base for the analysis was the model developed by
Leech and Short and briefly introduced by Semino (2004). The aim of the analysis was to find out
which of the categories of speech presentation are most frequently used in the three different text types
and the corpus as a whole, which speech act verbs are most commonly used when interpreting
someone’s speech in fiction and non-fiction texts, and last but not least, to explain our statistical
findings.The analysis shows that the most commonly used speech presentation mode is DS which
enables the author to make his characters seem independent of the narrator. Any move away from the
DS end of the scale brings with it the feeling of narrator interference. IS and NRSA, which are
moderately common in the investigated text types, feel like forms which are heavily controlled by the
narrator, because the words used are not those of the characters. The choice of forms of speech
presentation as well as the choice of speech act verbs has significant consequences for the degree of
importance that is attributed to different utterances, for the projection of viewpoint, and the creation of
a contrast between the characters. The different forms of speech presentation account for interesting
and effective variations in manipulating the narrative point of view in order to achieve a wide range of
significant effects. In particular, in fiction or news reports they can bias and control the readers’
empathy towards the characters.
Key words: reporting speech, reporting thought, direct and indirect speech, corpus, analysis, fiction,
academic prose, newspaper.
1. Introduction
Reporting is important because a huge amount of information we get comes from
other people. The nature of the language of reporting has long been of interest to a range of
scholars and has been the subject of numerous and thorough studies throughout the last 40
years, which investigate how someone else’s words and thoughts are mediated by the
narrating voice or text. The study of the character speech and thought presentation is an
important aspect of the narrative discourse. It has been extensively investigated within
stylistics, syntax as well as narratology21
. Moreover, research in this area has been also done
by philosophers22
, applied linguists23
, conversation analysts24
and psychologists25
, who
21
D. Cohn, G. Leech, M. Short, S. M Fludernik, E. Semino, and S. Rimmon-Kennan 22
Clark and Gerrig, 1990 23
Buttny, 1997; Thompson, 1996; Baynham and Slembrouck, 1999; Myers 1999 24
Holt 1999
112
examined in great detail the forms and functions of reported discourse (Jeffries, 2010). One of
the most widely accepted frameworks for the description of the phenomenon in this tradition
is Leech and Short’s (1981) model, which has been developed on the basis of the analysis of a
corpus of written narratives with the help of other scholars over a number of years and refined
by investigating the nature of speech, thought and writing presentation (henceforth ST&WP)
in spoken, as opposed to written, data (McIntyre, 2003).
2. Speech and thought presentation
Speech and thought presentation investigates how a speaker or writer presents the
speech or thought of other people. There is a range of ways in which writers and speakers can
present others' words or thoughts, and the choices made are important in determining what
sort of impression the reader or listener will get of the character whose speech or thoughts are
being presented. The writer/speaker can express the illocutionary force of an utterance by
using not only performative verbs but also other reporting signals (nouns, adjectives,
adjuncts) to affect the meaning of the original utterance. For example, the disagreement of a
politician with some plan: “No... We think the changes we have put forward are right.” can be
interpreted as: His stubborn response dismayed fellow Tory MPs..., which intensifies the
negative context or purpose of the original utterance.
There have been various classifications of speech presentation. The most elementary
concepts, which are presented in most pedagogical grammars, are those of direct and indirect
speech. There is a fundamental difference between direct and reported (indirect) speech. The
essential semantic difference between these two modes is that when one uses direct speech to
report what someone has said one reproduces the words used verbatim, whereas in indirect
report one expresses what was said in one’s own words by maintaining the general idea of the
utterance. In fact, while the two mentioned are the main modes of reporting, they are only 2 of
a larger number of possibilities. Besides the categories of Direct Speech (DS) and Indirect
Speech (IS) that a writer has available, according to Leech, Semino and Short (2004, 2007),
one also has the possibility of representing speech using Free Indirect Speech (FIS), a
Narrative Report of a Speech Act (NRSA), or a Narrative Report of Action (NRA). These
categories vary in the degree of control posited by the character or the narrator, representing a
progression from the most faithful to the least faithful to the speech act alone (DS,
25
Ravotas and Berkenkotter 1998
113
FIS,IS,NRSA,NRA). The following are a set of options for the reporter from which he will
choose according to what is most suitable for his communicative purpose.
DS – Direct speech
The original speech in inverted commas with a reporting clause.
He said, “I have been there several times.“
FIS – Free indirect speech
Representation of the original speech with no reporting clause retaining some features of
DS such as direct questions and vocatives.
He disagreed with their proposal.
IS – Indirect speech
Report in the words of the reporter applying backshifting of the original version
(changes in tense, pronouns and adverbials of time and place to align with the time of
reporting) Reporting clause + reported clause (subordinate NC)
He told me that he had been there several times.
NRSA - Narrator’s report of speech act
Performative Verb +NP/PP
She refused the offer. Tom accused him of lying.
NRA - Narrator’s report of action
VP+PP or NP
He talked about his experience. They presented the project.
The representation of a character's thought is essentially the same as that for the
representation of speech. The writer can represent grammatically a character’s thoughts
through Direct (DT) or Indirect Thought (IT), Free Indirect Thought (FIT), a Narrative Report
of a Thought Act (NRTA), or a Narrative Report of a Thought/an Act (NRT).
DT – Direct thought
He thought “I can't do it myself”.
FIT - Free indirect thought
He couldn't do it himself.
IT – Indirect thought
He thought that he couldn't do it himself.
NRTA - Narrator's report of thought act
He considered the likelihood of giving up.
114
NRT - Narrator's report of thought/an Act
He thought about asking for help.
However, there is a noticeable difference between the effects created by speech and
thought presentation categories. G. Leech and M. Short (2007, p. 344) state that the reason for
this is the difference between what is considered a norm for speech presentation categories on
the one hand, and thought presentation categories on the other. Whereas the baseline for
speech presentation is the direct speech (DS), the norm for the thought presentation is not
direct thought (DT) but its indirect form (IT), as it comes closest to acknowledging the fact
that it is implausible to suggest that we can directly observe the thoughts of others. Thought is
not “verbally formulated and so cannot be reported verbatim” (Leech and Short, 2007). We do
not actually have direct access to other people's thoughts and feelings by not claiming to
correspond precisely to the thinker's precise thoughts (http://www.languageinconflict.org/the-
world-through-language/presenting-speech-and-thought.html).
Speech and thought presentation can be a problematic area of investigation, especially
when it comes to discourses which can’t be fitted into traditional categories of direct vs.
indirect speech and thought. There is a long tradition focussing on speech and thought
presentation in written fiction (Banfield, 1973; McHale, 1978; and Fludernik, 1993). In
fictional discourse (literary texts), no anterior speech event actually take place, everything is
invented by the author. Some recent works on non-literary texts (Roeh and Nir, 1990;
Slembrouck, 1992; Caldas-Couldhard, 1994 and Waugh, 1995) have highlighted the need for
more systematic accounts of ST&WP phenomena across a wider range of text types (M.
Wynne, M. Short, and E. Semino 1998). To analyse the presentation of someone’s though in
non-literary texts is even more difficult. Thoughts often relate to information, beliefs or
attitudes that the characters could not express openly, and therefore tend to contrast with what
the characters do say (Cohn, 1978; Semino and Short, 2004).
3. Data and methodology
The main aim of our study is to draw attention to some important aspects of four
different text types, namely discourse presentation (the ways of interpreting other’s words)
and the ways the reported message is expressed. The object of interest lies in exploring the
choices, patterns and variations in the presentation of characters’ words in different text types.
It is also of interest to what extent the original speech changes when it is reported and how
115
linguistic choices might affect the projection of point of view and the potential for readers’
sympathy towards the characters, as well as the readers’ perceptions of the characters.
The variants of speech activities will be categorised according to Leech and Short’s
(1981) model of speech and thought presentation. We will draw our attention especially to IS,
FIS and NRSA, which are in the middle of the scale of faithfulness to the original speech and
examine the speech act verbs that the writers use help to perceive the nature of the described
event more intensively; thus these verbs to a considerable extent influence the reader and
enhance the pragmatic function of the report.
The investigated material is a small size corpus of authentic, naturally occurring,
connected, contextualized interactional language of different texts of written English (a
collection of authentic texts comprising three text types: academic prose, newspaper articles,
and fiction, all representing contemporary British and American English). The various texts
compiling the subcorpora chosen for our investigation were randomly selected regardless the
media, the topics, the level of language, and the age, sex, or social group of the users. The
analysed texts within the subcorpora are of different length, but each subcorpus is
approximately of the same breadth (15,000 words). The whole corpus thus includes more than
45,000 words. The texts chosen for the analysis are similar to one another as for the topic, but
differ not only in terms of grammar but also in terms of style, and degree of formality.
The subcorpus of academic prose consists of research papers, academic articles, and
passages from textbooks taken from different websites and study fields, such as psychology,
medicine, economics, linguistics, etc. The analysed newspaper columns were sampled from
across the various topics found in most newspapers. The texts covered the following major
areas: domestic news, foreign/world news, arts (including cinema, theatre, fine arts, fashion,
etc.), and social news (including reports about society people, environment, crime, etc.). The
last investigated corpus, which is contemporary fiction, consists of several chapters taken
from five different books: Brown, D. Inferno (2013); Patterson, J. Run for your Life (2009);
Hislop, V. The Return (2008); Bond, A. The Truth About Ruby Valentine (2011); Connelly,
M. Nine Dragons (2009). It is assumed that there exists an informal-formal continuum
between the texts as they are taken from different genres and printed sources. Both the formal
and functional description of the analysed structures will offer an overall view on different
structural types/ways of presenting others’ speech and the faithfulness to the original in the
report.
In order to find out the different ways of presenting speech, we applied quantitative
analysis of the different texts within the investigated corpus focusing particularly on the way
116
(structures) in which characters’ utterances are presented. It is a well-known fact that news
reports provide the reader with opinions, ideas, judgements, and feelings of the people
involved, they present the attitude of the writer/author or the editorship; moreover, they
enable the author to select and transform, it was interesting to investigate how the choice of
particular structures affects the projection of point of view and the potential for readers’
sympathy towards the characters, and to trace how faithfully the words by others are
presented. We were also interested in the extent to which the mentioned structures occur in
the texts of different degree of formality as it is supposed that the language used in reports in
different printed media differs. According to D. Crystal (1997), it can be accounted for the
fact that particular papers tend to adjust the language of their articles to the needs and
expectations of their readers. First of all we distinguished between presentation of speech and
thought. The structures that did not involve any of the mentioned presentations and those that
we considered as ambiguous were not analysed. Then we analysed each text in terms of the
use of structures which were marked as speech presentation and compared the results with
that of the whole corpus. In the process of unfolding analysis the following research questions
have emerged:
1/ Which of the categories of speech presentation are most frequently used in the corpus as a
whole and in three different text types?
2/ Which performative (speech act) verbs are most commonly used when interpreting
someone’s Speech in fiction and non-fiction texts?
4. Analysis and discussion
Since the selection and transformation takes place when the writer chooses and
introduces verbalisations, it was interesting to analyse the ways of reporting the original
events with regard to the selected grammatical structures and formulations/expressions of the
participants of the original communication event.
It is evident that different forms of speech presentation represent different degrees of
faithfulness to the original words; therefore, they cannot be regarded only as syntactic forms
of the same proposition. Especially, the category of FIS, defined both by lexical and syntactic
features and where the marker of subordination is absent, is a very loose concept (Leech and
Short, 2007). In non-literary texts, where ideology and persuasion is of high interest, and the
text writers effort is to manipulate the reader’s viewpoint, it might be difficult to find out
“whose voice one is hearing in processing a text as we may be misled about the words that
117
someone uttered or be persuaded because of who we think uttered the words” (Jeffries, 2010,
p. 144).
In our analysis, first of all we searched the corpus for all instances of presentation;
afterwards, we distinguished between presentation of speech and thought. The structures that
appeared ambiguous were not involved in the analysis. Some sentences included more than
one category, and there were a lot of sentences that did not involve speech or thought
presentation at all.
Table 1: The proportion of speech and thought presentation in each text type
and the whole corpus
Text type Speech presentation Thought presentation
Fiction 62.5% 37.5%
Newspapers 73.8% 26.2%
Academic prose 80.3% 19.7%
whole CORPUS 72.2% 27.8%
Table 1 above shows that there is almost three times as much speech as thought
presentation within the whole corpus. Looking at the text types separately, the incidence of
speech presentation (SP) is higher than that of thought presentation (TP) in each of them;
however, the proportion of SP to TP differs. The big difference in the proportion of speech
presentation to thought presentation in academic prose can be attributed to the fact that
academic writing requires exact articulation and identification of the speaker’s/writer’s
position in relation to the issues. The authors of this type of discourse do not present beliefs,
thoughts, or attitudes that the characters could not express openly, but present persuasive
arguments supporting their position, and identify the relevance of their argument. To do that,
they make use of communication verbs that express verbal processes or report information,
such as argue, claim, conclude, say, state, suggest, show, prove, demonstrate etc. as in the
example Thompson-Schill argue that generalized “selection demands” increase in complex
sentences, potentially confounding the signal for grammatical processing…
When analysing how speech is presented, we focused on the following categories as
introduced by Semino (2004):
DS (Direct speech) – the exact words in the report represented by utterances enclosed
within quotation marks and an independent reporting clause. The language used is appropriate
118
to the speaking character, e.g. The spokesman also said: “The strike may have resulted in
collateral damage to a nearby medical facility.”
FIS (Free indirect speech) – resembles indirect speech in shifting tenses and other
references, represented by utterances without a reporting clause using language that is partly
appropriate to the narrator (tense and pronouns) and partly to the speaking character
(grammar, lexis, deixis). In this technique the character's voice is partly mediated by the voice
of the author; in other words, it retains some features of direct speech (direct questions and
vocatives), e.g. Stefanie wondered whether we should not have driven straight home.
IS (Indirect speech) – conveys the report in the words of the reporter, represented by
utterances with a reporting clause followed by a subordinated reported clause. The language is
appropriate to the narrator with verbs generally ‘backshifted’ in tense and changes in
pronouns and adverbials of time and place are made to align with the time of reporting. In
indirect speech the reported clause does not re-enact the original speech event, but gives
rather current speaker’s perspective on the original sayer’s words. For instance, the sentence
of direct speech “So what are you going to do about the Friday dance class?” asked Maggie.
Could be rendered in indirect speech as: Maggie wanted to know what I was going to do about
the Friday dance class. The current speaker in the reported clause is referred to as I and not as
you as in direct speech.
NRSA (Narrator’s report of speech act) – expressed by a speech act verb usually
followed by a noun or prepositional phrase representing the subject matter of the speech act,
e.g. Stokes rejected Afghan claims that Taliban fighters had been in the hospital grounds.
NRA (Narrator’s report of action) - is a verbalization process which can be followed
by a prepositional or a noun phrase with minimal reference to speech taking place, e.g. She
replied to him.
To get the answer to our first research question, at first, it was necessary to find out
which stylistic variant (direct or indirect) was used to present speech in the different text
types. Then, indirect presentations of speech were put into categories based on their
faithfulness to the original version, starting with the most faithful to direct speech (DS), which
is free indirect speech (FIS) through indirect speech (IS), narrator’s report of speech act
(NRSA), and to narrator’s report of action (NRA). Table 2 below shows the results of
quantitative analyses of speech presentation in the different investigated text types within the
corpus.
Table 2: The distribution of different categories of speech presentation in the corpus
119
Speech presentation DS FIS IS NRSA NRA Total
Fiction
303
60.8%
45
9%
57
11.4%
72
14.5%
21
4.3%
498
51.9%
Newspapers
126
34.9%
24
6.7%
141
39%
62
17.2%
8
2.2%
361
37.6%
Academic prose
9
8.9%
11
10.9%
51
50.5%
27
26.7%
3
3%
101
10.5%
Whole CORPUS
438
45.6%
80
8.4%
249
25.9%
161
16.8%
32
3.3%
960
100%
The quantitative analysis of the five categories of speech presentation show to what
extent the investigated texts differ in terms of the frequency of particular speech presentation
variants. The division into DS, FIS, IS, NRSA, and NRA indicate that that Direct speech (DS)
is by far the most frequent form of speech presentation. In our analysis it accounts for 45.6%
of all speech presentation in the corpus and is highly represented especially in fiction.
However, it is not the most frequently occurring category in each of the investigated text
types. A significant cline can be noticed from the lowest figure of only 9 instances (8.9%) in
academic texts to the highest, 303 (60.8%), in fiction, so on this level it can be claimed that
the use of DS is text type relevant. In newspapers DS occupies the second place, and in
academic prose it belongs to the least frequently used category. The figures in Table 1 above
suggest that the occurrence of direct speech depends on the degree of formality of the text
type: the greater the frequency of DS, the less formal the text is.
The results indicate that indirect speech (IS), realized by reporting clauses followed by
subordinated reported clauses, is the second most commonly used variant of reporting other
people’s speech. It represents 25.9% of reported speech within the whole corpus. Here again,
there is an evident cline from its lowest incidence in fiction texts (11.4%) followed by
newspapers (39%) to the highest in academic prose (50.5%). In contrast to DS, which is very
common in less formal text types, the relatively high incidence of IS indicates that it is more
typical for formal texts.
Narrator’s report of speech act (NRSA) in total makes up 16.8%, which is the third
most commonly occurring way of speech presentation. The distribution of this category
120
within particular text types differs. While in fiction and academic prose NRSA is the second
most frequently used way of presenting speech, in newspapers it occupies the third position
being used less commonly than DS and IS. In fiction it accounts for 14.5% and it is more
frequent that IS. With 80 occurrences, which is 8.4% in the whole corpus, free indirect speech
(FIS) belongs to the least frequent form of speech presentation together with Narrator’s report
of voice (NV), which is the most minimal with only 3.3% in the corpus. Both are relatively
infrequent in all three investigated text types.
4.1 Direct Speech
The most frequently occurring category of speech presentation in the subcorpora of
fiction and newspapers is direct speech (DS). This form of speech presentation is “associated
with the faithful, verbatim reproduction of an original utterance” (Semino, 2004), the actual
words uttered by the speaker. It consists of an independent clause/s in inverted commas and
conveys the illocutionary force of speech act/s, its propositional content. Direct speech
includes all the deictic properties (tense, pronouns, lexis) appropriate to the speaker. The use
of DS in fiction and news reports helps to foreground particular situations, events and
dramatize the narrative as in: “Robert, do you realize what this means? It means the words
cerca trova were already in your subconscious! Don’t you see? You must have deciphered this
phrase before you arrived at the hospital! You had probably seen this projector’s image
already…but had forgotten!” (Brown, 2013). Using DS the writer draws attention to the
importance of the situation and according to Semino (2004, p. 8) “gives the reader the
impression that he/she is listening directly to the character’s voices without mediating
interference of the narrator”.
In news reports using DS foregrounds the actual words of speakers, especially
politicians, so that the reader feels he/she gets what the speaker himself said, not a version
worded through the journalist’s language, as in: “I can’t say there is big optimism,” said one
of the diplomats. “There is a strong feeling that protection of external borders is the
competence of the countries. Very many want to state their reservations”
(http://www.theguardian.com, cit. 16/10/2015). DS is conventionally associated with the
faithful, verbatim reproduction of an original utterance. In serious non-fictional writing there
may be a greater reluctance to use the direct forms where the original cannot be reproduced.
However, DS can sometimes be misused in order to affect many people's views. The direct
discourse forms journalists use may be accompanied by expressions which undermine the
121
reliability of the actual words uttered by the speaker, or show the speaker in a bad light, e.g.
Guy Verhofstadt, leader of Liberal group Alde in the European Parliament, yesterday warned
the EU: „From a cost, security and cultural perspective, this is completely the wrong move“.
(http://www.dailymail.co.uk, cit. 12/10/2015). Indeed, it is likely that readers will have
different expectations of the status of material within quotation marks depending on their
perception of the text-type they are reading (Short, Semino, Wynne, 2002). As Bayman and
Slembrouck (1999) suggest, the use of DS in news reports fails to represent a range of aspects
of spoken utterances, such as voice quality, accent, intonation, paralinguistic features, etc. The
English writing system is bad at representing these features of speech, but journalists can, and
obviously do make use of conventional orthographic representations to suggest
pronunciations, or insert information into the narrative description by the use of adverbs or
speech act verbs as in “Plan Frontera Sur has turned the border region into a war zone,”
argued Alberto Donis convincingly (http://www.theguardian.com, cit. 16/10/2015). This is
even more common in fiction where the use of adverbs and reporting signals helps the writer
express the way the actual words were pronounced as well as the illocutionary force of the
original utterance, e.g. “Sienna Brooks,” the man declared suddenly, the words crystal clear.
(Brown, 2013), or “Ruby?” she yelled.” Is that you?” (Bond, 2011). Short, Semino and
Wynne (2002) argue that fictional examples are fundamentally different from discourse
presentation in non-fictional language use, given that they do not, by definition, refer back to
an independent antecedent occasion when the language being re-presented was originally
produced.
The use of direct speech in the subcorpus of academic prose is minimal with only 9
instances, which is by far the least compared with that of fiction and newspapers. This can be
attributed to the character of academic writings, which is a product that results from thinking
and lacks face-to-face interactiveness. The example: In a recent review, Kaan and Swaab
(2002) note: “Broca’s area shows increased activity not only to contrasts such as….” Is by
its form it is DS; nevertheless, it is clear from the context that it is in fact presentation of some
writing.
4.2 Free Indirect Speech
Free indirect speech (FIS) is the category in between DS and IS. It is thought of as a
freer version of an indirect form. It is the mixture of features, some of which are appropriate
for IS, and some for DS in that it resembles IS in shifting tenses and other references. As a
122
result it is often very ambiguous as to whether it represents faithfully the words of the
character or whether it is the narrator's words which are being used. In FIS the reporting
clause is omitted, which allows the reported clause to take on some of the syntactic
possibilities of the main clause, and in this respect share some of the features typically
associated with DS (direct questions and vocatives). Leskiv (2009, p. 52) suggests that FIS
captures something between speech and thought which can neither be paraphrased in a
propositional form nor cast into an expression with a new first person referent. A prototypical
example of FIS is Sienna wondered what was wrong with his face (Brown, D. Inferno, 2013).
The context indicates that it is a representation of something Sienna says and verb ‘wonder’
here may be interpreted as a query rather than a representation of her thought.
As Semino and Short (2004) state, free indirect speech (FIS) is considered to be the
least frequent category of speech presentation, yet “it is believed to be linguistically more
complex than other forms since it is a mixture of direct and indirect features it be lexical,
grammatical or deictic markers of subjectivity” (Semino and Short, 2004). The figures
obtained from the analysis of our corpus support this finding and also suggest that this way of
speech presentation with only 8.4% occurrence is not very common in literary and non-
literary texts.
4.3 Indirect Speech
Indirect speech (IS) the second most frequently used category in the corpus with 249
instances, which represent 25.9% of all types of speech presentation found in the investigated
texts. The figures show that IS in academic texts (50.5%) as well as in newspapers (39%) is
by far the most common form of presenting someone’s speech compared to the other four
categories. Out of 361 instances of reporting speech in newspapers, 141 use the form of IS,
which consists of a reporting clause with a finite verb expressing the speech act and a reported
clause, which is subordinated grammatically to the reporting clause with all the deictic items
appropriate for the narrator's 'voice' (Prosecutor Ulf Willuhn said officials would carry out a
psychiatric examination to investigate whether that was the man's primary motive or whether
his health played a role.). IS tends to be used where the content of what a character says is
more relevant, in context, than the form of the relevant utterance(s). The propositional content
of the original speech act is specified, but no claim is made to present the words and
structures originally used to utter that proposition (McIntyre et al., 2003)
123
The reported clauses are most commonly realised by post-predicate nominal that-
clauses and zero that-clauses. The post-predicate reported clauses are controlled by a wide
range of verbs falling into different semantic domains. The most frequently used verb in the
reporting clause is say as in US military officials said that deployment of some 3,000 troops
had begun for the three-month Operation Atlantic Resolve (http://www.theguardian.com)
together with other speech act and communication verbs such as admit, agree, announce
confirm, deny, inform, etc. In newspaper texts, though in a limited number, reported clauses
appear also after some other communication verbs (recommend, insist, suggest, and demand),
which are used to propose a potential course of action rather than report some information,
e.g. Mr. Butler rightly recommends that the JIC chairmanship should be the holder's last job
(The Independent on Sunday, cit. 27/09/2015).
The lowest number of speech presentation as a whole was found in the analysed
academic texts. In the subcorpus comprising about 15,000 words, there were only 101
instances of speech presentation to be found. Academic texts usually present procedures and
convey referential information that characterizes most commonly inanimate entities in the
text. Since academic writing is the product that results from thinking, it makes much more use
of thought presentation (which enable the writers identify their attitudes in relation to the
discussed issues within the field) than presentation of speech. There is also a difference in the
use of verbs controlling reported clauses in academic prose compared with that in the other
text types. The most commonly occurring communication verbs in academic prose are
assume, argue, suppose, indicate, conclude, prove and suggest. These verbs enable the writers
to present persuasive arguments supporting their position and identify the relevance of their
argument, e.g. Fodor suggests that in this area less research needs to be done. The relatively
high occurrence of IS compared to the frequency of DS in academic texts can be attributed to
the character of this text type and its level of formality.
IS in fiction is not as common as in newspapers and academic writings. Its occurrence
is even lower than that of NRSA. It is due to the fact that in this genre there is a relatively
strong preference for fictional dialogues presented by DS; moreover, there is an extremely
high frequency of mental verbs to describe the thoughts and other cognitive states of fictional
characters used in narratives, so the majority of reported clauses report rather thought than
speech. The most common speech act verb controlling reported clauses in our subcorpus is the
verb say and tell, less common but still moderately used are admit, agree, complain, deny,
explain, mention, suggest, etc. as in Wearily Kelly explained once more that she should like to
see or speak to Max. (Bond, 2011).
124
4.4 Narrator’s report of Speech Act and Narrator’s report of Action
The figures in Table 2 show that narrator’s presentation of speech act (NRSA), which
accounts for 16.8% of all instances of speech presentation in the whole corpus is more
common than FIS but less frequent than IS. If we look at its incidence in the different text
types, the frequency of occurrence of NRSA is higher than that of IS only in fiction. NRSA is
considered to be more indirect than indirect speech, the narrator is in significant control of the
report.
Fictional NRSA is realized in sentences which consist of brief references to the
illocutionary force of utterance that signals that a speech act has occurred with no focus on
what was said nor what words were uttered as in “After that he barked out a list that
demanded accurate answers” (Hislop, 2009). NRSA, reporting a minimal account of the
original statement, is used to summarise stretches of discourse (Leech and Short, 2007), e.g.
“He criticised Ella’s behaviour.” (Bond, 2011). As a consequence, NRSA often has a
backgrounding function, i.e. it tends to be associated with relatively unimportant utterances,
as compared with more direct forms of presentation.
In academic prose, NRSA is the second most frequently occurring form of speech
presentation. It might be due to the fact that in this type of non-fiction texts there is a
relatively strong preference for communication verbs such as demand, explain, ensure, report,
suggest, and show, which are used in the structure VP+ NP/PP as in “Halgren suggested also
other kinds of computational system.” (Pinker, S. So How Does the Mind Work?, 2005),
which is typical for scientific and research papers.
The least frequently used category of speech presentation in the whole corpus in
general as well as in each of the three analysed text type is narrator’s report of action (NRA)
with only 32 instances of 960, which represents only 3.3%. This category that does not refer
to the original speech at all and where the narrator totally controls the report has the highest
occurrence in fiction. It is sometimes rather difficult to distinguish between NRA and pure
narration as in “Bosh turn to her and whispered something.” (Connelly, 2009).
5. Conclusion
In conclusion, speech presentation categories present a wide variety of possibilities for
writers. These modes exhibit various features that can be exploited by fiction and non-fiction
writers in a number of ways. Table 2 gives an overview of the frequency distribution of
125
different categories of speech presentation in the whole corpus and each subcorpus (text type)
within it. As seen above, the most commonly used speech presentation mode is DS which
enables the author make his characters seem independent of the narrator. Any move away
from the DS end of the scale brings with it the feeling of narrator interference. IS and NRSA,
which are moderately common in the investigated text types, feel like forms which are heavily
controlled by the narrator, because the words used are not those of the characters. The
different forms of speech presentation account for interesting and effective variations in
manipulating the narrative point of view in order to achieve a wide range of significant
effects. In particular, in fiction or news reports they can bias and control the readers’ empathy
towards the characters.
In the course of the analysis we focused on the choice of speech act verbs used in the
indirect categories of speech presentation in different text types. We found out that the verb
wonder is the most commonly used verb in FIS in each text type. By far the most common
communication verb found in IS in the corpus is the verb say. Tell and ask are much more
frequent in fiction than in newspapers and academic prose. Verbs such as argue and suggest
were to be found relatively equally distributed in all three text types. Except for these, fiction
makes use of agree, admit, complain, deny, explain, and mention, the most commonly used
verbs in newspaper text are announce confirm, deny, inform and assume, suppose, and
conclude in academic writings. As for the most frequently used verbs to report speech act
(NRSA) in fiction, these are: admit, agree, complain, deny, explain, mention, and suggest.
The verb demand is the most frequent in newspapers and explain, ensure, report, suggest, and
show in academic texts. The choice of forms of speech presentation as well as the choice of
speech act verbs has significant consequences for the degree of importance that is attributed to
different utterances, for the projection of viewpoint, and the creation of a contrast between the
characters.
References
Banfield, A., 1973. Narrative style and the grammar of direct and indirect speech. Foundations of
Language 10.
Baynham, M., and Slembrouck, S., 1999. Speech representation and institutional discourse. Text, 19
(4), 61-81.
Biber, D., 1988. Variation across Speech and Writing. Cambridge: Cambridge University Press.
126
Biber, D., Johansson, S., Leech, G., Conrad, S., and Finegan, E. 1999. Longman Grammar of Spoken
and Written English. Pearson Education Limited.
Bond, A., 2011. The Truth About Ruby Valentine. Penguin Books, Ltd.
Brown, D., 2013. Inferno. London: Bantam Press.
Caldas, C., and Couldhard, M. eds., 1994. Texts and Practices. London: Routledge.
Conelly, M., 2009. Nine Dragons. Orion Books.
Crystal, D., 1997. The Cambridge Encyclopedia of the English Language. 2nd
ed. Cambridge:
Cambridge University Press.
Fludernik, M., 1993. The Fictions of Language and the Language of Fiction. London: Routledge.
Hislop, V., 2008. The Return. Headline Publishing Group, UK.
Jeffries, L., 2010. Critical Stylistics. The Power of English. Palgrave McMillan.
Leech, G., and Short, M., 2007. Style in Fiction. 2nd
ed. Harlow: Pearson Longman, 255-281.
Leskiv, A., 2009. The literary phenomenon of free indirect speech. Studia Anglica Resoviensia, 6.
McHale, B., 1978. Free Indirect Discourse: A Survey of Recent Accounts. Poetics and Theory of
Literature.
McIntyre, D., Bellard-Thomson, C., Heywood, J., McEnery, A., Semino, E. and Short, M., 2003. The
construction of a corpus to investigate the presentation of speech, thought and writing in written and
spoken British English. In: D. Archer, P. Rayson, A. Wilson and A. McEnery, eds. Proceedings of the
Corpus Linguistics 2003 Conference. Lancaster University: UCREL Technical Papers, 16, 513-523.
McIntyre, D., 2003. Building a corpus to investigate the presentation of speech, thought and writing in
spoken British English. PALC (Practical Applications in Language Corpora) Conference 2003, Łodz
University.
Patterson, J., 2009. Run for your Life. Arrow Books, UK.
Roeh, I., and Nir, R., 1990. Speech presentation in the Israel radio news: ideological constraints and
rhetorical strategies, Text, 10.
Semino, E., 2004. Representing characters’ speech and thought in narrative fiction: a study of England
by Julian Barnes. Academic Journal Style, 38 (4), 428.
Semino, E., and Short, M., 2004. Corpus Stylistics: Speech, Writing and Thought Presentation in a
Corpus of English Writing. London: Routledge.
Short, M., Semino, E. and Wynne, M., 2002. Revisiting the notion of faithfulness in discourse
presentation using a corpus approach 1. Language and Literature, 11(4), 325-355.
Tannen, D., 1982. Oral and literate strategies in spoken and written narratives. Language,
58 (1), 1-21, Linguistic Society of America, available at: http://www.jstor.org/stable/413530.
127
Wynne, M., Short, M. and Semino, E., 1998. A corpus-based investigation of speech, thought and
writing presentation in English narrative texts. In: A. Renouf, ed. Explorations in Corpus Linguistics.
Amsterdam: Rodopi, 231-245.
http://www.languageinconflict.org/the-world-through-language/presenting-speech-and-thought.html